Impact of the Vertical Variation of Cloud Droplet Size on the Estimation of Cloud Liquid Water Path and Rain Detection

Abstract Cloud droplet effective radius (DER) and liquid water path (LWP) are two key parameters for the quantitative assessment of cloud effects on the exchange of energy and water. Chang and Li presented an algorithm using multichannel measurements made at 3.7, 2.1, and 1.6 μm to retrieve a cloud DER vertical profile for improved cloud LWP estimation. This study applies the multichannel algorithm to the NASA Moderate Resolution Imaging Spectroradiometer (MODIS) data on the Aqua satellite, which also carries the Advanced Microwave Scanning Radiometer (AMSR-E) for measuring cloud LWP and precipitation. By analyzing one day of coincident MODIS and AMSR-E observations over the tropical oceans between 40°S and 40°N for overcast warm clouds (>273 K) having optical depths between 3.6 and 23, the effects of DER vertical variation on the MODIS-derived LWP are reported. It is shown that the LWP tends to be overestimated if the DER increases with height within the cloud and underestimated if the DER decreases with height within the cloud. Despite the uncertainties in both MODIS and AMSR-E retrievals, the result shows that accounting for the DER vertical variation reduces the mean biases and root-mean-square errors between the MODIS- and AMSR-E–derived LWPs. Besides, the manner in which the DER changes with height has the potential for differentiating precipitative and nonprecipitative warm clouds. For precipitating clouds, the DER at the cloud top is substantially smaller than the DER at the cloud base. For nonprecipitating clouds, however, the DER differences between the cloud top and the cloud base are much less.